Applying fuzzy thesis on the selection of e-learning ...



A Fuzzy Selection Model of E-Learning System Developers

Huey-Ming Lee1), Chih-Jen Hsieh1), Mu-Hsiu Hsu2), Tsung-Yen Lee3)

1) Department of Information Management, Chinese Culture University

55, Hwa-Kung Road, Yang-Ming-San, Taipei(11114), TAIWAN

2) Taiwan Knowledge Bank Co., Ltd., 3F, NO.60, Po-Ai Road, Taipei(10041), TAIWAN

3) Department of Securities, Bank of Overseas Chinese

4TH FL., No. 102, Hern-Yang Road, Taipei, TAIWAN

Abstract: - The purpose of this study is to propose a selection model of e-learning system developers by fuzzy sets theory. Via using this model, we can not only select the qualified e-learning system developers but also respond to the selected developers’ professional talents.

Key-Words: - E-learning system developers; Grade of ability

1 Introduction

With the advent of the broadband network and the life-long learning era, being professional in e-learning has become a trend and also part of the necessary strategies in advanced education [1]. Therefore, the selection and promotion of e-learning system developers has had an influential impact on the strategy of e-learning content and the industry’s talent cultivation.

Qualifications vary due to the different job responsibilities of e-learning system developers. Therefore, when it comes to evaluating or selecting qualified e-learning system developers, we not only have to consider their professional ability, but also the various potential talents and abilities of the developers. At the moment, the criteria for choosing a developer are still evaluated on two aspects, either good or not good. However, this is not always very objective and is somewhat vague. That is why this study quotes Lee et al. [5] in which they mentioned using an indefinite evaluation model to select qualified e-learning system developers.

Many factors are considered in building up a selection model, because each factor has its different ability level. Therefore, firstly we have to classify the most important categories; and then divide the different levels of each category, and we proceed with the indefinite evaluation of each developer in each category to decide his ability level. After that, we are able to choose the best qualified e-learning system developers.

2 The Attitudes Demanded of E-Learning Developers

From the Taiwan Knowledge Bank’s experience of e-learning system development, they have decided that the basic attitudes a qualified e-learning system developer should have seven towards, saying, team work, continuous learning, professional ability, research and development, leadership, loyalty, and work attitude.

Keast [3] has presented five issues in his research focused on e-learning development at innovated management. Husmann and Miller [2] have mentioned twenty-four issues in a integrated model that discussed the e-learning requirements, and we have chosen eleven professional evaluations and combined the attitudes into leadership and continuous learning. Spencer and Spencer [7] have presented twelve issues in an evaluation chart of IT software professional staffs that discussed the professional talent requirements, and we have chosen seven professional evaluations and divided them into three categories. Botturi [1] considered knowledge as a relationship and showed how this provides helpful insights in instructional design and education management, that professional e-learning developers should have six basic talents which are combined into professional ability and creativity research and development. Keller [4] has submitted five hundred and thirty-one persons of science and engineer in research into the relation of work involvement, and divided the relations of loyalty into three categories.

We have realized that an e-learning system developer’s professional ability combines many factors. Different jobs have different responsibilities. Therefore, qualified professional knowledge and ability is the only way to be a successful e-learning system developer candidate, which means that we have to have complete and objective evaluation as our principle to choose and decide on the right developer candidate.

3 Fuzzy Assessment Model

3.1 Hierarchical structure model

When selecting the e-learning system developers, the considerable terms and qualifications should include many aspects. Also, to present the selection conditions in a systematic way, we apply the analytic hierarchy process (AHP) [5, 6] to conduct a hierarchical structure model for selecting the e-learning system developers, as shown in Figure.1.

3.2 Grade of ability

When selecting the qualified developers, applying the selective items in Figure 1 as selection criteria, each item has its required ability level which means the candidates’ abilities. We ranged the grade of ability level into thirteen ranks as shown in Table 1 [5].

In many cases, we cannot express uncertainty problems simple by using the concept of probability. With the availability of concept of fuzzy sets theory, we can solve the problem under fuzzy circumstances. Moreover, fuzziness can be quantified by using the properties of fuzzy numbers.

Referring to Lee et al. [5], we made the linguistic values 0, 1, 2, …, 12 into corresponding reasonable fuzzy numbers with triangular membership functions as listed in Table 2 [5].

3.3 Evaluating the aggregative ability method

Let [pic]represent the weight of the attribute[pic], and [pic]denote the weight of the item[pic]as shown in Table 3, g(skj)represent the value which is derived by the defuzzfication by the centroid method of the grade of ability skj.

In referring the Lee et al. [5] algorithm, calculation steps are as follows:

Step 1. Let

[pic]

[pic]

for k=1, 2, …, 7, where n(k) is the number of evaluation items for attribute Xk, then we have n(1)=2, n(2)=5, n(3)=7, n(4)=3, n(5)=3; n(6)=4, n(7)=3; g(skj) is the value defuzzfied by the centroid method of the evaluated item Xkj; Wk,j is the weight of the item Xkj.

Step 2. The final ability of aggregative evaluation is by the centroid method as follows:

Let [pic]

[pic]

Then, the value of R1 is the aggregative ability of the e-learning system developer.

4 Numerical Examples

Suppose we have the weights, grade of ability for each evaluation item of the three developer candidates, saying A, B, and C, as shown in Table 4. The weights are fixed, which can not be affected by different candidates.

Final evaluation results can be obtained after calculation through above mentioned fuzzy assessment algorithm in Section 3.3. The total ability for each candidate shows in Table 5. To compare the total ability is B>C>A, candidate B is the best.

5 Conclusion

This study applies fuzzy sets theory on the model for selection of e-learning developers, which can enable the e-learning industry to adjust the category weights and each selective item’s importance with the actual need. Therefore, the selected candidates will be more professional and qualified.

Nevertheless, the selection model helps to define the candidates’ actual profession, and also serves as a reference of promotion for the e-learning employers.

Reference:

[1] L. Botturi, Knowledge as Relationship and E-learning, World Conference on E-Learning in Corp., Govt., Health., & Higher Ed., 2002(1), 144-153.

[2] D. E. Husmann & M. T. Miller, A holistic model for primary factors in the ecology of distance education course offerings, Journal of Distance Education, 11(1), 1996, pp.101-111.

[3] D. A. Keast, Toward an effective model for implementing distance education programs, The American Journal of Distance Education, 11(2), 1997, pp.39-55

[4] R. T. Keller, Job involvement and organizational commitment as longitudinal predictors of job performance: A study of scientists and engineers, Journal of Applied Psychology, Vol. 82, No.4, 1997, pp.539-45

[5] Huey-Ming Lee, Shu-Yen Lee, Tsung-Yen Lee and Jan-Jo Chen, A New Algorithm for Applying Fuzzy Set Theory to Evaluate the Rate of Aggregative Risk in Software Development, Information Sciences, Vol. 153, 2003, pp.177-197

[6] T. L. Satty, The Analytic Hierarchy Process, McGraw-Hill, New York, 1980.

[7] L.M. Spencer, & S.M. Spencer, Competency at Work, John Wiley & Sons, Inc., 1993, pp.107-173

Table 1 Linguistic value of grades of ability [5]

Thirteen grades of ability

0: Nil ability

1: Definitely low

2: Extra low

3: Very low

4: Low

5: Slightly unimportant

6: Middle

7: Slightly high

8: High

9: Very high

10: Extra high

11: Definitely high

12: Perfect ability

Table 2 Fuzzy numbers of thirteen

grades of ability [5]

|Grade of ability |Fuzzy number |

|0 |N0=(0.0,0.0,0.0) |

|1 |N1=(0.0,0.0,0.1) |

|2 |N2=(0.0,0.1,0.2) |

|3 |N0=(0.1,0.2,0.3) |

|4 |N4=(0.2,0.3,0.4) |

|5 |N5=(0.3,0.4,0.5) |

|6 |N6=(0.4,0.5,0.6) |

|7 |N7=(0.5,0.6,0.7) |

|8 |N8=(0.6,0.7,0.8) |

|9 |N9=(0.7,0.8,0.9) |

|10 |N10=(0.8,0.9,1.0) |

|11 |N11=(0.9,1.0,1.0) |

|12 |N12=(1.0,1.0,1.0) |

Table 3 Contents of selection model [5]

|Attribute |Weight |Selection |Weight |Grade of |

|(Xi) |(Wi) |item (Xij) |(Wij) |ability |

| | | | |(sij) |

| |W1 |X11 |W11 |s11 |

|X1 | | | | |

| | |X12 |W12 |s12 |

|X2 |W2 |X21 |W21 |s21 |

| | |X22 |W22 |s22 |

| | |X23 |W23 |s23 |

| | |X24 |W24 |s24 |

| | |X25 |W25 |s25 |

|X3 |W3 |X31 |W31 |s31 |

| | |X32 |W32 |s32 |

| | |X33 |W33 |s33 |

| | |X34 |W34 |s34 |

| | |X35 |W35 |s35 |

| | |X36 |W36 |s36 |

| | |X37 |W37 |s37 |

|X4 |W4 |X41 |W41 |s41 |

| | |X42 |W42 |s42 |

| | |X43 |W43 |s43 |

|X5 |W5 |X51 |W51 |s51 |

| | |X52 |W52 |s52 |

| | |X53 |W53 |s53 |

|X6 |W6 |X61 |W61 |s61 |

| | |X62 |W62 |s62 |

| | |X63 |W63 |s63 |

| | |X64 |W64 |s64 |

|X7 |W7 |X71 |W71 |s71 |

| | |X72 |W72 |s72 |

| | |X73 |W73 |s73 |

Table 4 Contents of the selection model of the example

|Attribute |Weight |Selection |Weight |Candidates A |Candidates B |Candidates C |

|(Xi) |(Wi) |item |(Wij) | | | |

| | |(Xij) | | | | |

| | | | |Grade of ability (sij) |Grade of ability (sij) |Grade of ability (sij) |

|X1 |0.1 |X11 |0.3 |6 |12 |10 |

| | |X12 |0.7 |10 |2 |9 |

|X2 |0.2 |X21 |0.3 |2 |10 |0 |

| | |X22 |0.2 |10 |9 |6 |

| | |X23 |0.1 |4 |11 |8 |

| | |X24 |0.3 |7 |10 |12 |

| | |X25 |0.1 |6 |11 |5 |

|X3 |0.2 |X31 |0.1 |9 |3 |1 |

| | |X32 |0.2 |6 |10 |5 |

| | |X33 |0.1 |8 |8 |4 |

| | |X34 |0.1 |9 |0 |6 |

| | |X35 |0.1 |6 |12 |6 |

| | |X36 |0.1 |8 |12 |7 |

| | |X37 |0.3 |5 |6 |11 |

|X4 |0.1 |X41 |0.4 |6 |10 |12 |

| | |X42 |0.3 |3 |12 |5 |

| | |X43 |0.3 |4 |1 |9 |

|X5 |0.1 |X51 |0.4 |5 |10 |3 |

| | |X52 |0.2 |4 |1 |2 |

| | |X53 |0.4 |6 |0 |12 |

|X6 |0.1 |X61 |0.2 |5 |10 |6 |

| | |X62 |0.2 |6 |11 |6 |

| | |X63 |0.3 |3 |8 |4 |

| | |X64 |0.3 |7 |3 |9 |

|X7 |0.2 |X71 |0.4 |5 |10 |3 |

| | |X72 |0.2 |4 |1 |2 |

| | |X73 |0.4 |6 |0 |12 |

Table 5 Aggregative ability value for each

Candidate

|Candidate A |Candidate B |Candidate C |

|0.536 |0.66556 |0.62762 |

Figure 1 Hierarchical structure model of e-learning system developer for aggregative ability

-----------------------

Aggregative ability

X7 Loyalty ability

X72 Work participation

X73 Personal character

Selection item Xij

Attribute Xi

X24 Central knowledge

X71 The meaning of work

X63 E-learning teaching

X62 E-learning training

X6 Continuously learning ability

X61 Service department

X5 Research and development ability

X4 Teamwork ability

X3 Leadership ability

X52 Application software operation quality

X53 Basic quality of Internet teaching application

X42 Communication & negotiation

X43 Relationship establishment

X51 IT course professional quality

X22 External Situational Analysis Ability

X33 Marketing Segmentation Ability

X34 Learning Activity & Process Management

X35 Evaluating Requirement Ability

X36Evaluative ability

X41 Teamwork

X37 Marketing ability

X25 Teacher course design for professional attitude

X31 Internal Situational Analysis Ability

X23 Common knowledge

X12 Initiative

X2 Professional ability

X22 Technical Expertise

X21 Analytical Thinking

X11 Impact and Influence

X1Work attitude ability

X64 E-learning experience

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download